15:01:55 #startmeeting horizon 15:01:56 Meeting started Wed Apr 22 15:01:55 2020 UTC and is due to finish in 60 minutes. The chair is amotoki. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:01:57 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:01:59 The meeting name has been set to 'horizon' 15:02:06 hi 15:02:08 #chair e0ne 15:02:09 Current chairs: amotoki e0ne 15:02:13 hi 15:02:13 hi 15:03:16 let's start 15:03:56 hi 15:03:57 I would like to cover ussuri-related topic first and let e0ne lead the remaining if needed 15:04:07 #topic announcements/notices 15:04:41 This week is the hardstring freeze for ussuri. 15:05:05 we already cut stable/ussuri last week 15:05:24 so we don't have much on this. 15:05:48 what we need to do is to review stable/ussuri reviews as usual. 15:06:04 the master branch is now open for victoria. 15:06:56 amotoki: can we get this one https://review.opendev.org/#/c/721674/ backported to Ussuri? it's a bugfix with new management command? 15:07:08 I'm sorry, didn't file a bug yet :( 15:07:38 e0ne: it just depends on when it lands. I think it is a normal bug fix. 15:08:05 amotoki: ok. thanks. I'll propose a chery-pick once it will be merged to master 15:08:10 e0ne: perhaps we will have another release from stable/ussuri to import translations so it would be part of the ussuri release. 15:08:42 any other announcements? 15:09:01 I've created an ehterpad for PTG planning 15:09:03 #link https://etherpad.opendev.org/p/horizon-v-ptg 15:09:31 we should request a virtual room for ptg this week if we need it 15:09:40 it looks like 'v' means both 'virtual' and 'victoria' :) 15:09:45 I'm going to do it today 15:09:50 amotoki: :) 15:10:29 which timeslots do you plan to request? 15:11:07 it depends on whole the team 15:11:29 I'm in UTC+3 timezone 15:11:50 I would like to have a good time for the all cores 15:12:04 #link http://lists.openstack.org/pipermail/openstack-discuss/2020-April/014126.html 15:12:19 and some small slot to cover US timezone in case if we need to cover some cross-project dicsuccions 15:12:33 timeslots are categorized into three zones. 15:13:03 #link https://ethercalc.openstack.org/126u8ek25noy 15:13:53 I'm OK for 13-21 UTC 15:14:01 I am okay either with 13-17UTC and 4-8UTC slots 15:14:09 21:00 UTC is pretty late, so it's not my favorite time:) 15:14:24 vishalmanchanda, rdopiera: what about yuo? 15:14:53 I am fine with 13-21 UTC slot. 15:15:17 there is no such slot "13-21UTC" :p 15:15:35 oops:). 15:15:49 you can check the ethercalc link above 15:15:58 i am fine with any slot. 15:16:09 any time slot work for me:). 15:16:28 go ahead. 15:17:09 :) 15:20:55 it looks like we can move on 15:20:59 #topic ussuri 15:21:05 e0ne: I don't have a preference 15:21:17 #undo 15:21:18 Removing item from minutes: #topic ussuri 15:22:23 13-17UTC is converted into 10pm-2am in my local time, but I think it fits most folks in horizon. 15:22:49 hopefully it can finish 16UTC but it is not a big problem to me. 15:23:55 anything else? 15:23:58 I think 2-3 hours will be enough 15:24:24 +1 15:24:35 +1 15:24:51 if we cannot cover all, perhaps we can reserve on-demand slots 15:25:32 +1 15:26:13 let's create a list of topics to discuss https://etherpad.opendev.org/p/horizon-v-ptg before we request some timeslot 15:28:09 yeah, I will dump items in my mind this week 15:29:24 amotoki: +1 15:29:47 moving on 15:29:56 #topic ussuri release 15:30:14 we already covered most things for ussuri release above 15:30:38 great work! 15:30:59 tempest-horizon plugin 1.0.0 was also releaed https://review.opendev.org/#/c/721500/ 15:31:15 it is cycle-automatic 15:31:44 v.1.0.0! 15:31:46 :) 15:31:53 hehe 15:32:22 it is 1.0.0 but I am not sure we would like to continue it or merge tempest tests into the integration tests :p 15:33:18 :) 15:34:05 it can be visit in V cycle. we run devstack only for one test. 15:35:26 :) 15:35:45 it's useful test to verify deployment with openstack services 15:35:56 apart from that, I just started a follow-up work on the community goal "Contributor and PTL guide" 15:36:29 it would be nice if we can finish it till the U release. 15:37:41 if you have bug fixes good for ussuri, feel free to backport them and ping us. 15:38:02 that's all from me on ussuri 15:39:04 amotoki: thanks a lot for leading us during this release! 15:39:33 yw. you all helped me a lot and we are the team. 15:40:12 #topic on-demand agenda 15:41:32 I would like to share two gate breakages we had last week for FYI 15:41:45 :( 15:41:58 the one is pytest usage broke all plugin UTs 15:42:26 I'm sorry for that:(. I broke plugins gates again :( 15:42:36 pytest availability is checked now to make pluginUTs work without pytest https://review.opendev.org/720844 15:42:54 amotoki: thanks for the fix. 15:43:01 perhaps we need to introduce a minimum cross gate with some horizon plugins 15:43:15 +1 15:43:22 amotoki: I want to raise this topic last week but i completely forgot:(. 15:43:28 it't in my TODO list for a while 15:43:52 it would be something like what requirements repo does. 15:44:16 the other thing is related to the migration to unittest.mock. 15:44:20 +1 15:44:37 oops. I broke everything again 15:44:51 amotoki: what was the issue with mock? 15:44:59 horizon migration to unittest.mock broke horizon plugin UTs which use create_mocks from the helper function. 15:45:32 create_mocks sets up mocks using unittest.mock and some other mocking use the mock library 15:45:57 oh.. :( 15:46:09 I haven't investigated the failure in detail but some inconsistency (mixed usage of unittest.mock and mock) caused the failure 15:46:23 it was observed in heat-dashboard 15:47:26 As far as I checked, it does not mean all plugin UT using create_mocks fail. 15:47:42 one example failure is https://zuul.opendev.org/t/openstack/build/1abdd087265648b68026a7f2ea6d590c 15:48:17 this is another reason that we need cross gate with horizon plugins 15:48:33 it is hard to detect such by human reviews 15:49:12 that's all I would like to share here. 15:49:31 I have a question to ask about pytest 15:49:39 amotoki: thank for pointing us in these topics 15:50:07 we had a lot of gate breakage in this cycle..... 15:50:29 django-babel, pyscss, unitest.mock, pytest..... 15:50:38 might had more 15:51:26 vishalmanchanda: go ahead. what is your question on pytest? 15:51:39 I have noticed some of plugins adds pytest in lower-constraints.txt and test-requirements.txt which fix their gate at that time. 15:51:49 So i have also purpose a patch in a plugin https://review.opendev.org/#/c/720652/ 15:52:14 amotoki: Should i revert it now? 15:52:29 it is up to plugin team 15:52:46 ok. 15:52:49 pytest requirement was an accident 15:53:09 if a plugin would like to use pytest for test runner, they can continue to use it. 15:53:47 agreed. 15:54:05 if not, it would be nice to clean up it so that requirements files have no unnecessary stuffs. 15:54:35 I have no preference on changes you made. 15:54:45 amotoki: ok i will check with that team. 15:55:24 thanks. 15:56:44 anything else? 15:56:56 nothing from me 15:56:57 nothing from my side. 15:57:26 let's close the meeting. thanks for joining. 15:57:31 take care yourselves 15:57:32 bye 15:57:34 #endmeeting